Measuring forest recovery at Orokonui Ecosanctuary

This is part 2 of a 3-part series highlighting our work testing the effectiveness of fenced sanctuaries for biodiversity conservation.  You can read part 1 here.

It has taken longer than I wanted, but I’ve now found some time to revisit our work at Orokonui from earlier this year.  We were re-measuring permanent vegetation monitoring plots as part of a one-year project, primarily supported by a British Ecological Society Research Grant.  As mentioned previously, our aim has been to test whether eco-sanctuaries restore habitat for threatened fauna.

Our work has involved 52 permanent 10 m x 10 m plots randomly located throughout the 307 ha Orokonui Ecosanctuary outside of Dunedin, New Zealand.  The plots were established between 2005 and 2007, so many had to be re-located.  That can prove quite challenging in the dense New Zealand bush:

Plot margin marked out by a yellow tape leading into a tree fern Cyathea smithii full of spiders!

Plot margin marked out by a yellow tape leading into a tree fern (Cyathea smithii) that was full of spiders!

But our efforts – led by our collaborator Kelvin Lloyd (Wildland Consultants, Landcare Research) have proved very successful.  We’ve managed to re-locate and measure 50 of the original 52.  One plot was destroyed to build an enclosure for Sirocco, our old friend from Maud Island who occasionally visits Orokonui, while another was cleared for the pest proof fence surrounding the property.  In addition to the plots inside of the fence, we have also measured 23 control plots located in nearby forests similar to that at Orokonui, dominated by broadleaved trees and kānuka.

A fairly large buffer needs to be created alongside the main fence.  This is used as a track inside the ecosanctuary.

A fairly large buffer needs to be created alongside the main fence. This is used as a track inside the ecosanctuary.

Once we relocate the corners of a plot – we quickly establish the plot margins.  We then take a series of detailed inventory measurements.  This first involves measuring the diameter at breast height (approx 1.35 m height) on all woody plants greater than 2 m tall.  We then count all the woody plants less than 2 m tall in 0-10 cm; 10-50 cm, 50-100 cm, and 100-200 cm height classes.  This can be quite consuming as some plots contain literally thousands of saplings and seedlings (the record being 2,860).  But censusing these is critical for measuring the success of the ecosanctuary.  We are hoping that eliminating seed predators, namely introduced rats and mice, is allowing more native tree seedlings to establish and survive.  Finally, the abundance of non-woody species (e.g. ferns, grasses, herbs) and lianes is assessed in four 1 m × 1 m subplots in each plot corner.  

Kelvin hard at work measuring seedlings.

Kelvin hard at work measuring seedlings into the dark.


A quick look through the data is revealing three key changes in the forest:

(1)  Succession is creating more bird habitat.

A large portion of Orokonui is covered in the advanced stages of early-successional kānuka forest.  This is now starting to transition to more closed-canopy late-successional forest dominated by large podocarps and broadleaved trees.  The transition can take decades.  But it is important because many native bird species rely on large trees.


Change in basal area of the 50 forest plots over time. Most points sit above the 1:1 line, indicating greater basal area in 2013/2014.

Kākā are just one of many bird species that need large broadleaved trees.  One reason that is not widely reported – despite plenty of anecdotal evidence – is that they feed on sugars produced by trees.  They do this by stripping away the outer bark of trees to access the phloem.


Broadleaved tree ring-barked by kākā beaks in Orokonui.

Larger trees, more of which are dominated by broadleaves, are therefore beneficial for birds such as kākā.


(2)  Densities of trees <2 m tall are increasing.

Seedling densities are increasing over time, most dramatically in plots with fewer big trees (i.e. less canopy cover).  Some of the changes are most noticeable for the large seeded trees that would have been preferentially consumed by rodents.  One such species is the lovely-named stinkwood Coprosma foetidissima, which also has among the largest seeds of the trees in Orokonui.

There is a 95% probability that densities of Coprosma foetidissima (10-200 cm tall) have increased by at least 100 trees per ha since the last time the plots were measured.  This is one of the species that responded strongest immediately after rats were eradicated on an offshore island in Fiordland.  Densities of trees <2 m tall are also much higher inside of the ecosanctuary than in nearby forests:


Coprosma foetidissima is doing a lot better inside of Orokonui than in the surrounding landscape.

Here, we’ve considered those trees in the 0-10 cm height tier as well because we are comparing across plots in the same year.  It makes less sense to include these trees when comparing among years because this height tier is generally an ephemeral reflection of inter-annual variability in seed production.

Regeneration of broadleaved trees should be good for birds.


(3)  Conservation is protecting the best forest in the region.

Forests located outside of Orokonui are failing to regenerate.  These areas are under tremendous pressures from a whole suite of introduced mammals that reach from the forest floor to canopy.  Rodents predate seeds and seedlings, ungulates browse saplings, and possums slowly defoliate and kill large canopy trees.  The net result is that there can be little left over in these forests that isn’t actually eaten!

Differences in the size structure of broadleaved trees inside and outside of the ecosanctuary reveal these stark differences.  This is just one example for the large broadleaved species commonly known as “broadleaf“.  What we find is that trees are recruiting to larger size classes over time, and there is an abundance of trees in small size classes, inside of the ecosanctuary.  But the species is virtually absent in the greater region, despite the very similar kānuka forest type.

The very palatable Griselinia littoralis is missing in similar kānuka-broadleaf forests outside of the ecosanctuary.

Palatable Griselinia littoralis is missing in similar kānuka-broadleaf forests outside of the ecosanctuary.  Inside the ecosancturary, large trees (diameter > 50mm) are becoming more abundant.

Again, this suggests that the ecosanctuary is protecting special habitat that will be beneficial to birds that need big trees and structure.



That’s it for now.  I hope to have more time to analyse the data in the coming months.  So in my final blog on Orokonui, I’ll consider the future of ecosanctuaries as a conservation intervention more broadly in an international context.


Male tui (Prosthemadera novaeseelandiae) drinking from one of the feeders in Orokonui.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s